Pollination by fungus gnats and associated floral characteristics in five families of the Japanese flora
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Annals of Botany 121: 651–663, 2018
doi:10.1093/aob/mcx196, available online at www.academic.oup.com/aob
Pollination by fungus gnats and associated floral characteristics in five families
of the Japanese flora
Ko Mochizuki* and Atsushi Kawakita
Center for Ecological Research, Kyoto University, Otsu, Shiga 520-2113, Japan
*For correspondence. E-mail: ko.mochizuki@ecology.kyoto-u.ac.jp
Downloaded from https://academic.oup.com/aob/article-abstract/121/4/651/4781678 by guest on 22 May 2020
Received: 6 November 2017 Editorial decision: 13 November 2017 Accepted: 13 December 2017
Published electronically 29 December 2017
• Background and aims Pollination by fungus gnats (Mycetophilidae and Sciaridae) is uncommon, but is nevertheless
known to occur in 20 genera among eight angiosperm families. Because many fungus gnat-pollinated plants possess
a dark red floral display, we hypothesized that fungus gnat pollination is more widespread among plants with similar
floral display than currently known. We thus studied the pollination biology of flowers with dark red pigmentation in
five families, focusing particularly on plants having small, flat, actinomorphic flowers with exposed nectaries and short
stamens, because these floral characteristics mirror those of a known fungus gnat-pollinated genus (Mitella).
• Methods We observed daytime and night-time floral visitors for a total of 194.5 h in Aucuba japonica
(Garryaceae), Euonymus spp. (Celastraceae), Disanthus cercidifolius (Hamamelidaceae), Micranthes fusca
(Saxifragaceae) and Streptopus streptopoides (Liliaceae). Visitors were categorized into functional groups, and a
pollination importance index (PII) was calculated for each functional group based on visitation frequency, pollen
load and behaviour on flowers.
• Key results Fungus gnats were dominant among the 1762 insects observed (36–92 % depending on the plant
species) and were the most important pollinators among all plants studied (PII: 0.529–1). Fungus gnat visits
occurred during the daytime and, more frequently, at dusk. Most often, pollen grains became clumped on the
ventral side of the head and/or thorax as the short-proboscid fungus gnats foraged on nectar and came into contact
with anthers located close to the flower base.
• Conclusions Pollination by fungus gnats is probably more common than previously thought, especially in
habitats similar to those of the plants studied (moist forest understorey, streamside or subalpine meadow) where
fungus gnats are abundant year-round. Our results further suggest that there may be a previously unnoticed
association between fungus gnat pollination and dark red coloration, and a shared overall floral architecture among
the plants studied.
Keywords: Aucuba japonica, dark red flower, Diptera, Disanthus cercidifolius, Euonymus, forest environment,
fungus gnat, Micranthes fusca, Mycetophilidae, pollination syndrome, Sciaridae, Streptopus streptopoides.
INTRODUCTION on mushrooms, fungal mycelia in rotten wood, decaying
plant material and, more rarely, bryophytes or other insects
Dipteran insects are among the most common visitors to angio- (Vockeroth, 1981; Steffan, 1981; Søli et al., 2000; Jakovlev,
sperm flowers and often play key roles as pollinators in many 2011, 2012). Fungus gnats are usually considered as inef-
plant communities (Kearns, 2001; Larson et al., 2001; Ssymank fective pollen carriers due to their small bodies, inconstancy
et al., 2008; Lefebvre et al., 2014; Inouye et al., 2015; Orford and weak flight ability (Mesler et al., 1980; Proctor et al.,
et al., 2015; Ollerton, 2017). However, the importance of dip- 1996; Larson et al., 2001; Willmer, 2011). Nevertheless,
teran pollination is possibly still underestimated, because pollination by fungus gnats is known to occur in eight
research continues to uncover plants that depend strongly on angiosperm families: Orchidaceae, Liliaceae, Asparagaceae,
dipterans for pollination, some of which attract dipteran pollina- Araceae, Aristolochiaceae, Polygonaceae, Apocynaceae and
tors by previously unsuspected mechanisms (Stökl et al., 2011; Saxifragaceae (Vogel, 1973; Jones, 1974; Ackerman and Mesler,
Ren et al., 2012; Song et al., 2014; Oelschlägel et al., 2014; 1979; Mesler et al., 1980; Sugawara, 1988; Fuller, 1994; Vogel
Heiduk et al., 2016). Small dipteran insects, such as fungus and Martens, 2000; Goldblatt et al., 2004; Okuyama et al., 2008;
gnats, gall midges and drosophilid flies, are particularly difficult Yamashiro et al., 2008; Barriault et al., 2010; Duque-Buitrago
to observe in the field, and thus their roles as pollinators are et al., 2014; Ollerton et al., 2017; Suetsugu and Sueyoshi, 2017).
probably not fully appreciated (Larson et al., 2001; Goldblatt Some of these plants are believed to mimic fungal oviposition
et al., 2004; Woodcock et al., 2014). sites because pollinator eggs are often observed on the flowers
Fungus gnats, or flies in Mycetophilidae, Sciaridae and (Vogel, 1978; Sugawara, 1988; Scanlen, 2006), whereas others
related Sciaroidea families, are small, weak-flying insects com- employ sexual deception (Blanco and Barboza, 2005; Phillips
mon in moist forest and riparian habitats. Their larvae feed et al., 2014), offer nectar reward (Ackerman and Mesler, 1979;
© The Author(s) 2017. Published by Oxford University Press on behalf of the Annals of Botany Company.
All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
652 Mochizuki and Kawakita – Fungus gnat pollination and shared floral characters
Mesler et al., 1980; Goldblatt et al., 2004; Okuyama et al., and small, with dark red petals (Figs 1B and 2A, B). Aucuba ja-
2004) or even sacrifice developing ovules to seed-feeding pol- ponica is one of the first woody plants to flower in early spring
linator larvae to accomplish the pollination (Song et al., 2014). (March–May).
Despite the diversity of mechanisms used by these plants to The spindle tree genus Euonymus comprises about 130 spe-
attract fungus gnats, several fungus gnat-pollinated plants share cies with a broad distribution range, including North America,
a characteristic floral display: petals or spathes with dark red Europe, Asia, Australia and Madagascar (Ma, 2001). Most spe-
pigmentation (Supplementary Data, Table S1). Dark red flow- cies are small trees or shrubs growing in a variety of habitats
ers are otherwise common in sapromyiophilous plants (Fægri including the forest edge, forest floor and rocky coast. The
and van der Pijl, 1979; Willmer, 2011; Johnson and Schiestl, flowers are small and typically greenish or yellowish; how-
2016; Shuttleworth et al., 2017) and are also known in sexu- ever, roughly 30 species have red, purple or chocolate-coloured
ally deceptive orchids (Gaskett et al., 2017) and some fly- and flowers (Ma, 2001). To date, 18 Euonymus species have been
beetle-pollinated plants (Young, 1985; Sakai and Inoue, 1999; recorded in Japan. These are typically small trees occurring
Downloaded from https://academic.oup.com/aob/article-abstract/121/4/651/4781678 by guest on 22 May 2020
Teichert et al., 2012; Policha et al., 2016; Luo et al., 2017); on forest edges or within forests, whereas some are shrubs or
however, knowledge of an association between such a floral shrublets on the forest floor. Four endemic species, E. lanceo-
display and any particular pollination system remains frag- latus, E. melananthus, E. tricarpus and E. yakushimensis, have
mentary. This observation motivated us to examine the floral red flowers and prefer the shaded forest floor or, occasionally,
visitors of dark red-flowered Aucuba japonica (Garryaceae) subalpine/boreal meadows. Inflorescences of Euonymus are
in 2014. We found that A. japonica flowers were heavily vis- cymes of up to ten flowers, developing on leaf axils. The flower
ited by fungus gnats and noticed that the flowers of Aucuba possesses a prominent nectary disc that surrounds the pistils
resemble those of the known fungus gnat-pollinated genus and stamens. Filament length varies among species; however,
Mitella (Okuyama et al., 2004) in several ways: they are small species with dark red flowers invariably have a very short to
(5–8 mm in diameter), flat and actinomorphic, with exposed almost no filament.
nectaries, short stamens and dark red floral displays. We thus Euonymus lanceolatus is an evergreen dwarf shrub that grows
examined if plants in other families with floral characteris- in shaded forests on the Sea of Japan side of the Japanese main-
tics similar to those of Mitella and Aucuba are also fungus land, where there is heavy snow in winter. The plants are 0.5
gnat-pollinated. m high and have flexible stems that are pressed flat against the
We searched for plants that share the above set of floral char- ground under snow (Ma, 2001; Fig. 1C). During June and July
acteristics in the Japanese flora. Consequently, we studied the (rainy season), each plant produces a few to dozens of flowers
pollinators of the following six shrubs and herbs: Aucuba ja- that open downwardly.
ponica (Garryaceae), three Euonymus species (Celastraceae), Euonymus melananthus is a small, deciduous shrub (up
Disanthus cercidifolius (Hamamelidaceae) and Micranthes to 1 m) growing on the shaded riparian forest floor, having a
fusca (Saxifragaceae) (Figs 1 and 2). Additionally, we studied wide distribution in cool–temperate areas of Japan, but occur-
Streptopus streptopoides (Liliaceae) because, although its flow- ring only infrequently. The plants resemble lianas, growing
ers are mostly greenish and have only small areas of dark red from chinks in the rocks along mountain streams. The flowers
pigmentation toward the tepal base, the overall floral morph- are dark red and 7–9 mm in diameter, blooming from June to
ology is very similar to those of Mitella, suggesting the pos- July (Fig. 1D). In one of the study populations (Mt Haku-san),
sibility of fungus gnat pollination (Figs 1 and 2). Based on a E. melananthus and E. lanceolatus grow in close proximity, and
3-year field study, we report that these seven studied plants are E. melananthus flowers earlier than E. lanceolatus, although
predominantly pollinated by fungus gnats. Our findings point to their flowering periods partly overlap.
a possibility that a certain set of floral characteristics is associ- Euonymus tricarpus is a deciduous shrub up to 2–3 m in
ated with fungus gnat pollination. height and often found beneath the shade of other vegetation in
subalpine or boreal meadows in northern Japan. The plants pro-
duce hundreds to thousands of flowers (8–10 mm in diameter)
MATERIALS AND METHODS during June and July. Their flower colour is variable compared
to those of E. lanceolatus and E. melananthus, ranging from
Study materials and study sites
pink to dark red.
Disanthus cercidifolius (Hamamelidaceae) is a deciduous
Aucuba (Garryaceae) is a genus of evergreen dioecious shrubs shrub up to 2–3 m in height, often found growing along streams
comprising ten species distributed from the Himalayas to tem- in temperate forests in very limited areas of Japan and China.
perate East Asia (The Plant List, 2013). Aucuba is one of three Disanthus cercidifolius is one of two species in the genus
genera that constitutes the order Garryales, the other two being Disanthus; the other is the recently described Vietnamese spe-
predominantly wind-pollinated: Garrya (Garryaceae) and cies D. ovatifolius (Averyanov et al., 2017). Disanthus cercidi-
Eucommia (Eucommiaceae) (Dahling, 1978; Stull et al., 2015). folius produces flowers from November to December when
Unlike those of the latter two genera, Aucuba flowers possess the leaves have already turned red (Fig. 1E). Its flowers are
an open, prominent nectar disc, which is suggestive of an ento- hermaphroditic and self-compatible, and produce nectar on the
mophilous pollination system (Abe, 2001; Liston, 2003). The basal areas of the petals. Xiao et al. (2009) reported drosophi-
Japanese laurel A. japonica, a shrub up to 2 m in height, is lid and syrphid flies as pollinators of D. cercidifolius; however,
one of the most common components of the moist, shady floor their observations were conducted only during the daytime.
of temperate deciduous, temperate and subtropical evergreen Micranthes fusca (Saxifragaceae) is a hermaphroditic per-
forests in Japan (Fig. 1A). The flowers of A. japonica are flat ennial endemic to Japan, which grows on moist rocks along
Mochizuki and Kawakita – Fungus gnat pollination and shared floral characters 653
A B
Downloaded from https://academic.oup.com/aob/article-abstract/121/4/651/4781678 by guest on 22 May 2020
C D
E F G
Fig. 1. Habitats and inflorescences of the studied plants: (A) A. japonica in forest understorey; (B) inflorescences of A. japonica (male plant); (C) E. lanceolatus
on forest floor; (D) flowering branch of E. melananthus; (E) D. cercidifolius on streamside vegetation; (F) M. fusca on streamside rock in subalpine meadow; (G)
S. streptopoides under shade of bamboo grass.
mountain streams or in alpine vegetation. The plants produce The Japanese twisted-stalk Streptopus streptopoides (Liliaceae)
dark red flowers 5–8 mm in diameter on the approx. 30-cm is a small perennial up to 50 cm in height, which grows in the
thyrse from July to September (Fig. 1F). The flower com- shade of subalpine vegetation in central Japan (Fig. 1G). These
prises a conspicuous, fleshy floral disc surrounded by ten plants produce single downward-facing flowers beneath the
stamens. The stamens are initially spread radially on the flat leaves from June to July. The tepals are generally greenish
floral plane, which are bent upwardly as the anthers dehisce. but have dark red markings toward the centre of the flower.
654 Mochizuki and Kawakita – Fungus gnat pollination and shared floral characters
A B
C D E
Downloaded from https://academic.oup.com/aob/article-abstract/121/4/651/4781678 by guest on 22 May 2020
F G H
I J
5 mm
Fig. 2. Flowers of the studied plants: (A) A. japonica (male flower); (B) A. japonica (female flower); (C) E. lanceolatus; (D) E. melananthus; (E) E. tricarpus; (F)
D. cercidifolius; (G) M. fusca; (H) S. streptopoides; (I) Mitella koshiensis, which has previously been reported to be pollinated by short-proboscid fungus gnats;
(J) Mitella furusei, which has previously been reported to be pollinated by long-proboscid fungus gnats.
The nectary is located on the tepal base. Phylogenetic studies light when they were foraging on flowers. All flower visit-
have revealed that Streptopus is closely related to bee-pollinated ors were captured; however, diurnal visitors of E. lanceola-
Prosartes and fungus gnat-pollinated Scoliopus (Mesler et al., tus, D. cercidifolius, M. fusca and S. streptopoides were very
1980, 2010). sensitive to the presence of the observer, so some visitors to
Detailed information on the study sites and dates of data col- these species could not be captured. When visitors were not
lection are provided in the Supplementary Data (Table S2). captured, their number and taxa were recorded. Visitors were
captured mainly using a glass vial or aspirator, and some vis-
itors of E. lanceolatus and S. streptopoides were captured
using a small hand-made insect net for the reasons described
Flower visitors and pollen load analysis
above. Although the insect net facilitated the capture of sen-
We conducted field observations of flower visitors from sitive fungus gnat visitors, their body pollen was lost when
2015 to 2017 in Japan. Flower visitors were observed during the fungus gnats struggled within the small net. Therefore, we
the daytime and night-time from 0500 to 2200 h at peak flower- did not include pollen load data for visitors captured using the
ing for each species. For each observation, we selected three insect net.
to ten flowering individuals and made 5–10 min observations Flower visitors were identified to the family level, and
of each, in turn. For night-time observations, a flashlight was fungus gnats belonging to Mycetophilidae, which were the
used. Although there is a possibility that the lighting affected most important pollinators among the studied species (see
insect visitation, floral visitors were generally insensitive to Results), were further identified to the genus level. AlthoughMochizuki and Kawakita – Fungus gnat pollination and shared floral characters 655
Sciaridae may contain several fungus gnat genera, including and Inouye, 1993), which was then spread on a glass slide
Bradysia and Scatopsciara, we assigned them all as Sciaridae to count the stained pollen grains. Pollen grains of the host
spp. without separating them to morphospecies, due to the plants were distinguished by comparing them to a pollen ref-
difficulty of distinguishing genera and species based on ex- erence of the plants flowering at each study site, although in
ternal morphology. Fungus gnats were also sexed. To facili- most cases the body pollen originated from the plants studied.
tate interpretation of the results, flower visitors were classified We were not able to discriminate pollen grains of co-flower-
into functional groups based on their taxonomy, morphology ing E. lanceolatus and E. melananthus in the Mt Haku-san
and behaviour on flowers (Fenster et al., 2004). We estab- population. These two species shared Neoempheria sp. 1 as
lished ten functional groups: fungus gnat, crane fly, other a pollinator, so heterospecific pollen may have been included
Nematocera, hover fly, non-syrphid Brachycera, Coleoptera, in these counts.
ant, non-ant Hymenoptera, Lepidoptera and the remaining
short-proboscid insects, including Trichoptera, Neuroptera
Downloaded from https://academic.oup.com/aob/article-abstract/121/4/651/4781678 by guest on 22 May 2020
and Collembola. Hemipteran insects were not categorized
into functional groups because they were rare (only two indi- Nectar secretion
viduals were observed) and carried no pollen.
To assess the relative importance of floral visitors as pol- Nectar secretion seems to be important for the pollination
linators, the pollinator importance index (PII) of each func- of the seven studied plant species, because nectar feeding by
tional visitor group was calculated following Lindsey (1984), flower visitors was observed in all species. We cut branches
Hansen and Molau (1994), and Johnson et al. (2009). PII is and/or stems bearing flowers, and these cut plants were
based on four parameters: (1) relative abundance (A); (2) pol- kept in water-filled vials. Entire plant cuttings were put into
len-carrying capacity (PCC), expressed as the average number plastic bags to prevent nectar evaporation. After 24 h, nectar
of pollen grains carried; (3) host fidelity (F), expressed as the was collected using a 1-μL capillary, and nectar volume was
average proportion of the pollen load containing host pol- calculated based on the proportion of the length to which the
len; and (4) pollinator effectiveness (PE), the probability that capillary was filled with nectar to the entire capillary length.
the foraging behaviour of a given visitor results in pollin-
ation (Lindsey, 1984). Lindsey (1984) assigned each visitor
RESULTS
group with 11 PE values ranging from 0 to 1 at 0.1 intervals
based on insect size, behavioural and morphological fit with
the flower, and movement among flowers. Because a detailed Overall pollinator composition
analysis of PE is beyond the scope of the present study, we In all seven plants studied, dipteran insects were the predominant
assigned a PE value of 1 for visitors that contacted both the visitors (71–96 % of the individuals observed), although taxo-
stigma and the anther, whereas insects that contacted neither nomic composition was variable among species (Tables 1 and
anther nor stigma or contacted either anther or stigma only S3). Among the dipteran insects, fungus gnats (Mycetophilidae
were given a PE value of 0. Because A. japonica is dioecious, and Sciaridae) were the most frequent floral visitors (36–92 %,
some visitor species of A. japonica that were collected only a Tables 1 and S3, Fig. 3). Based on the PII scores, fungus gnats
few times were observed on either male or female plants only. were the most important pollinators for all studied species; the
We assigned PE values of 1 for such visitor species if they PII of fungus gnats ranged from 0.53 in A. japonica to 1 in
contacted the anthers (on male plants) or stigma (on female S. streptopoides (Fig. 4). The major fungus gnat taxa present
plants), because all other A. japonica visitors that were col- differed among plant species, although the genus Boletina was
lected on both male and female plants invariably touched the common (Table 2). Notably, the fungus gnats collected on flow-
anthers and stigma. ers were strongly female-biased (proportion of females: 0.79 in
As in Lindsey (1984), we calculated a pollination import- S. streptopoides to 0.96 in A. japonica, Table 2). However, no
ance value (PIV) for each visitor group by multiplying the fungus gnat visitors exhibited oviposition behaviour; instead,
four values: PIV = A × PCC × F × PE. We then gave each all fungus gnat visitors were observed feeding on nectar. All
functional group a PII score by calculating the relative pol- floral visitors, with the exception of moths, grass bugs and a
lination importance among all visitor groups: PII = PIV/ƩPII camel cricket, contacted both the stigma and anthers (Table 1).
(Lindsey, 1984; Hansen and Molau, 1994; Johnson et al., Most pollen grains carried by flower visitors originated from
2009). Separate PII scores were calculated for Mycetophilidae the plants at which the visitors were captured; however, crane
and Sciaridae. Although pollen load does not necessarily flies, an empidid fly, and dermestid beetles on A. japonica and
reflect true PE (Zych et al., 2013), and pollination efficiency various Brachycera flies and one mycetophilid on M. fusca car-
ideally should be evaluated by pollen removal, pollen depos- ried some heterospecific pollen (Table 1).
ition and seed set per visit (Spears, 1983; Mayfield et al., 2001;
Ne’eman et al., 2010; King et al., 2013), due to the logistical
difficulties of obtaining these data for a broad assemblage of
Pollinators of Aucuba japonica
visiting insects for seven plant species, we decided that PII
provides a basis for comparison that is better suited to the aim During the 33.5-h observation period, we recorded 465 insect
of our study. individuals of 71 species belonging to 40 families of six orders:
Pollen grains carried on the bodies of visitors were counted Diptera, Coleoptera, Lepidoptera, Hymenoptera, Trichoptera
under a stereo-microscope. When pollen grains formed a and Ephemenoptera (Tables 1 and S3). Dipteran insects rep-
clump, they were removed using basic fuchsine jelly (Kearns resented 88.6 % of all visitors (Tables 1 and S3). Specifically,656 Mochizuki and Kawakita – Fungus gnat pollination and shared floral characters
Table 1. Importance of each functional group as a pollinator of the seven plant species studied.
Plant Functional group Individuals Average Proportion Relative Pollinator Pollination Pollination Pollen placement site
collected pollen of host abundance effectiveness importance importance
(indivuals load pollen (F) (A) (PE) value (PIV) index (PII)
with (PCC)
pollen)*
A. japonica
Fungus gnat 319 (220) 17.8 1 0.686 1 12.211 0.529
Mycetophilidae 110 (98) 27.8 1 0.237 1 6.576 0.285 Mainly head, thorax and
occasionally abdomen
Sciaridae 209 (122) 12.5 1 0.449 1 5.618 0.244 Head, thorax, abdomen
Crane fly 39 (35) 61.1 0.975 0.084 1 4.996 0.217 Mouthpart, ventral thorax
Downloaded from https://academic.oup.com/aob/article-abstract/121/4/651/4781678 by guest on 22 May 2020
Other Nematocera 29 (21) 14.3 1 0.062 1 0.890 0.039 Entire body
Non-syrphid Brachycera 25 (18) 23.3 0.946 0.054 1 1.185 0.051 Legs, head
Coleoptera 12 (11) 13.1 0.692 0.026 1 0.233 0.010 Elytra
Ant 10 (10) 10 1 0.022 1 0.215 0.009 Thorax, abdomen
Non-ant Hymenoptera 26 (21) 57.2 1 0.056 1 3.198 0.139 Entire body
Lepidoptera 2 (2) 3 1 0.004 1 0.013 0.001 Legs
Others (caddisfly, mayfly) 3 (1) 19.3 1 0.006 1 0.125 0.005 Thorax, abdomen
E. lanceolatus
Fungus gnat 80 (2/6) 29.5 1 0.615 1 18.154 0.991
Mycetophilidae 76 (2/2) 88.5 1 0.585 1 51.738 0.991 Foreleg femur
Sciaridae 4 (0) 0 – 0.031 1 0 0 –
Non-syrphid Brachycera 29 (8/18) 2.1 1 0.223 1 0.463 0.009 Head, abdomen
Ant 18 (3/4) 1.3 0 0.138 1 0 0 Head, abdomen
Others (camel cricket, grass bug) 3 (0) 0 – 0.023 0 0 0 –
E. melananthus
Fungus gnat 21 (13/17) 67.2 1 0.553 1 37.137 0.924
Mycetophilidae 16 (10/12) 87.8 1 0.421 1 36.947 0.895 Foreleg femur
Sciaridae 5 (3) 9 1 0.132 1 1.184 0.029 Foreleg femur
Other Nematocera 5 (0) 0 – 0.132 1 0 0 –
Non-syrphid Brachycera 5 (2) 1.2 1 0.132 1 0.158 0.004 Legs
Coleoptera 3 (1) 16.7 1 0.079 1 1.316 0.032 Thorax, head
Ant 1 (0) 0 – 0.026 1 0 0 –
Lepidoptera 2 (0) 0 – 0.026 1 0 0 –
Others (green lacewing) 2 (2) 31.5 1 0.053 1 1.658 0.040 Head
E. tricarpus
Fungus gnat 697 (189) 4.3 1 0.895 1 3.843 0.548
Mycetophilidae 31 (26) 39.0 1 0.040 1 1.564 0.223 Foreleg femur
Sciaridae 661 (163) 2.7 1 0.855 1 2.279 0.325 Foreleg femur
Crane fly 9 (6) 42.2 1 0.012 1 0.491 0.070 Legs (femur), head
Other Nematocera 21 (0) 0 – 0.027 1 0 0 –
Non-syrphid Brachycera 22 (12) 44.6 1 0.028 1 1.269 0.181 Head, thorax, abdomen
Coleoptera 12 (9) 30.5 1 0.016 1 0.473 0.067 Head, thorax
Ant 7 (4) 2.1 1 0.009 1 0.019 0.003 Thorax
Non-ant Hymenoptera 8 (4) 88.1 1 0.010 1 0.912 0.130 Head, thorax
Lepidoptera 2 (0) 0 – 0.003 0 0 0 –
D. cercidifolius
Fungus gnat 31 (19/28) 156.8 1 0.388 1 60.760 0.863
Mycetophilidae 26 (16/23) 156.0 1 0.325 1 50.700 0.720 Legs (femur), head
Sciaridae 5 (3) 160.4 1 0.063 1 10.025 0.142 Legs (femur), head
Crane fly 14 (8) 16.5 1 0.175 1 2.888 0.041 Legs (femur), head
Hoverfly 11 (7) 2 1 0.138 1 0.330 0.005 Mouthpart
Non-syrphid Brachycera 19 (11/17) 26.5 1 0.238 1 6.294 0.089 Head, legs, abdomen
Lepidoptera 4 (0) 0 – 0.050 0 0 0 –
Others: caddisfly 1 (1) 13 1 0.013 1 0.163 0.002 Head
M. fusca
Fungus gnat 121 (65/67) 24.9 0.98 0.708 1 17.277 0.841
Mycetophilidae 105 (57/59) 29.1 0.98 0.614 1 17.510 0.802 Legs, head
Sciaridae 16 (8) 9.2 1 0.094 1 0.865 0.040 Head, thorax
Other Nematocera 2 (0) 0 – 0.012 0 0 0 –
Hover fly 1 (1) 45 0.46 0.006 1 0.124 0.006 Head, thorax
Non-syrphid Brachycera 37 (27) 17.5 0.82 0.216 1 3.100 0.142 Legs, head
Coleoptera 2 (2) 17.5 1 0.012 1 0.210 0.010 Head, thorax
Non-ant Hymenoptera 8 (1) 0.8 1 0.047 1 0.038 0.002 Head, thorax
S. streptopoides
Fungus gnat 97 (13/21) 9 1 0.924 1 8.314 1
Mycetophilidae 93 (13/17) 10 1 0.886 1 8.857 1 Legs (femur)
Sciaridae 4 (0) 0 – 0.038 1 0 0 –
Other Nematocera 1 (0) 0 – 0.010 1 0 0 –
Non-syrphid Brachycera 2 (0) 0 – 0.019 1 0 0 –
Coleoptera 4 (0) 0 – 0.038 1 0 0 –
Others (grass bug) 1 (0) 0 – 0.010 0 0 0 –
* The single numbers in parentheses indicate the number of individuals that carried pollen out of the number of individuals collected. The two numbers separated
by a slash in parentheses donate the number of individuals that carried pollen (numerator) out of the number of individuals examined (denominator).Mochizuki and Kawakita – Fungus gnat pollination and shared floral characters 657
A B
Downloaded from https://academic.oup.com/aob/article-abstract/121/4/651/4781678 by guest on 22 May 2020
C D E
F G H I
Fig. 3. Fungus gnat visitors of the studied plants: (A) Boletina sp. (arrows) visiting a male inflorescence of A. japonica at dusk; (B) Sciaridae sp. (arrows) visiting
an inflorescence of E. tricarpus at dusk; (C) Mycetophila sp. visiting D. cercidifolius; (D) Boletina sp. visiting M. fusca; (E) Brevicornu sp. 2 visiting S. strep-
topoides; (F) Boletina sp. 1 collected on the flower of A. japonica. Pollen grains were attached mainly to the head and thorax. (G) An unidentified mycetophilid
fungus gnat collected on a E. melananthus flower, with massive pollen load on a femur of the forelegs (arrowhead). (H) Sciaridae sp. collected on a E. tricarpus
flower. Pollen grains were attached to the hairy femur (arrowheads). (I) Boletina sp. 7 caught on a D. cercidifolius flower, with numerous pollen grains on the
femur and mouthparts (arrowhead).
Mycetophilidae and Sciaridae fungus gnats were by far the Pollinators of Euonymus lanceolatus
most frequent visitors (319 individuals: 68.6 % of all visitors). During the 67.5-h observation period for the two E. lanceo-
Among the mycetophilids, Boletina was the dominant genus; latus populations, 130 insects of 12 species belonging to seven
Boletina sp. 1 was overwhelmingly more frequent than other families of four orders were recorded (Tables 1 and S3). The
mycetophilid species (Table 2). Flowers were visited by sciarid mycetophilid gnats Neoempheria spp. and drosophilid flies
fungus gnats throughout the day and by mycetophilid fungus were the predominant visitors at both study sites. Visitation
gnats at dusk (1830–1930 h, Fig. 3A). Although the most fre- by Neoempheria spp. occurred only during humid afternoons
quent visitors were various sciarid species, the numbers of pol- (1400–1600 h), although visitation was highly unpredictable
len grains carried were much lower than those of mycetophilids. throughout the observation period. Neoempheria spp. were
Pollen grains were attached mainly on the ventral side of the observed to forage for nectar. Drosophilid flies were observed
thorax and head, and occasionally on the abdomen (Fig. 3F). during the daytime, but many of these rested on the petals,
Pollen-laden fungus gnats were repeatedly observed on female abaxial surface of the flower or flower stalk. The numbers
plants, suggesting that fungus gnats are capable of transporting of pollen grains carried were much higher in Neoempheria
pollen between individuals. spp. than in drosophilid flies (Table 1), probably due to658 Mochizuki and Kawakita – Fungus gnat pollination and shared floral characters
A B C D E
F G
Downloaded from https://academic.oup.com/aob/article-abstract/121/4/651/4781678 by guest on 22 May 2020
Mycetophilidae Coleoptera
Fungus gnat
Sciaridae Ant
Crane fly Non-ant Hymenoptera
Other Nematocera Lepidoptera
Hover fly Others
Non-syrphid Brachycera
Fig. 4. Pie charts showing the pollination importance index (PII) of functional pollinator groups: (A) A. japonica; (B) E. lanceolatus; (C) E. melananthus; (D)
E. tricarpus; (E) D. cercidifolius; (F) M. fusca; (G) S. streptopoides.
behavioural differences on the flowers. Pollen grains attached Pollinators of Disanthus cercidifolius
to Neoempheria spp. were found mainly on the femurs of the
During the 18.5-h observation period, we recorded 80 insects
forelegs.
of 22 species belonging to 11 families of four orders (Tables 1
and S3). The flowers were visited diurnally by hoverflies and
flies of other brachyceran families, including Drosophilidae,
Pollinators of Euonymus melananthus Lauxaniidae and Tachinidae, which was consistent with obser-
vations made by Xiao et al. (2009) in China. However, the
During the 29.25-h observation period, we recorded 37 indi-
flies found in the current study generally carried a small pol-
viduals of 19 species belonging to 11 families of five orders
len load despite their large body sizes (Table 1). Around sunset
(Tables 1 and S3). During the daytime, flowers were occasion-
(1800–1900 h), flowers were frequently visited by fungus gnats
ally visited by mayflies and common flies including droso-
(Fig. 3C), which carried considerable amounts of pollen (up
philids; however, around sunset (1830–1930 h) the flowers
to 1154 grains; Fig. 4, Table 1). Pollen grains were attached to
were visited by fungus gnats. Mycetophilids, Myomya spp.,
the femurs of the forelegs and to the ventral side of the head
Neoempheria spp. and species of an unidentified genus were
(Fig. 3I). The dominant fungus gnat taxon was Boletina. All
observed. Fungus gnats carried pollen grains only on the hairy
visitors foraged on nectar secreted by the petals, except for hov-
femur (Fig. 3G). The numbers of pollen grains carried by fungus
erflies, which also foraged on pollen.
gnats were the highest among the functional groups (Table 1).
Pollinators of Micranthes fusca
Pollinators of Euonymus tricarpus
During the 14-h observation period, we recorded 171 insects
During the 17.25-h observation period, we recorded 779 of 31 species belonging to 19 families of three orders (Tables 1
individuals of 48 species belonging to 27 families of four orders and S3). Insect visits occurred both during the day and at dusk;
(Tables 1 and S3). Although E. tricarpus received visits by a however, the flowers were heavily visited by mycetophilids
much larger number of insects than the other two Euonymus at dusk (1730–1830 h). Mycetophilidae and Sciaridae fungus
species, the flower visitors were predominantly dipteran insects gnats were the most frequent visitors, and carried more pollen
(751 individuals, 96.4 %), the majority of which were fungus than other visitors except one rare hoverfly. Several mycetophi-
gnats (697 individuals; Fig. 3B, Table 1). Visitation by sciarid lid genera were recorded including Brevicornu and Exechia
fungus gnats was observed throughout the day, but visitations (Table 2). Pollen grains were attached on the ventral sides of
by mycetophilids were observed only around sunset (1800– the legs and head of fungus gnats (Fig. 3D, Table 1).
1900 h). Hymenopteran visitors, such as ichneumonoid wasps
and saw flies, carried the largest numbers of pollen grains, fol-
lowed by brachyceran flies, crane flies and mycetophilid fungus Pollinators of Streptopus streptopoides
gnats (Table 1). Although sciarid flies were frequent visitors,
they carried fewer pollen grains than did mycetophilid flies During the 14.5-h observation period, very few insect taxa
(Table 1). Fungus gnats carried pollen grains on the femurs of were observed visiting the flowers. The flowers were almost
the forelegs (Fig. 3H). exclusively visited by mycetophilid fungus gnats, BrevicornuMochizuki and Kawakita – Fungus gnat pollination and shared floral characters 659
Table 2. Taxonomic composition of the fungus gnat visitors.
Plant Fungus gnat family Species Individuals Female
collected ratio
A. japonica Mycetophilidae Boletina sp. 1 92 0.51
Boletina sp. 2 1 1
Boletina sp. 3 1 1
Brevicornu sp. 1 2 1
Coelosia sp. 1 1 1
Mycetophila sp. 1 2 1
Mycetophila sp. 2 2 1
Mycetophila sp. 3 1 1
Mycomya sp. 1 1 1
Downloaded from https://academic.oup.com/aob/article-abstract/121/4/651/4781678 by guest on 22 May 2020
Orfelia sp. 1 5 1
Saigusaia sp. 1 1 1
unidentified species 1 1
Sciaridae spp. 209 0.9
E. lanceolatus Mycetophilidae Epicypta sp. 1 2 1
Epicypta sp. 2 1 1
Epicypta sp. 3 1 1
Mycetophila sp. 4 1 1
Neoempheria sp. 1 13 0.46
cf. Neoempheria sp. 1 (not captured) 58 –
Sciaridae spp. 4 0.75
E. melananthus Mycetophilidae Mycomya sp. 2 1 1
Mycomya sp. 3 2 1
Mycomya sp. 4 1 1
Mycomya sp. 5 1 1
Neoempheria sp. 1 2 0
Neoempheria sp. 2 1 1
unidentified species 4 0.75
unidentified species (not captured) 4 –
Sciaridae spp. 5 1
E. tricarpus Mycetophilidae Aphrastomyia sp. 1 1
Boletina sp. 4 14 0.57
Boletina sp. 5 9 0.78
Brevicornu sp. 2 2 0
Brevicornu sp. 3 1 0
Epicypta sp. 4 1 0
Epicypta sp. 5 2 1
Mycetophila sp. 5 1 1
Sciaridae spp. 663 0.96
D. cercidifolius Mycetophilidae Boletina sp. 6 2 1
Boletina sp. 7 8 0.5
Exechia sp. 1 2 1
Mycetophila sp. 6 4 1
Mycetophila sp. 7 5 1
Rymosia sp. 2 1
unidentified species (not captured) 3 1
Sciaridae spp. 5 0.8
M. fusca Mycetophilidae Boletina sp. 8 1 1
Brevicornu sp. 2 27 0.59
Epicypta sp. 6 2 0.5
Exechia sp. 2 11 1
Mycetophila sp. 8 4 1
Mycetophila sp. 9 3 1
Mycetophila sp. 10 3 1
Orfelia sp. 2 7 0
Rymosia sp. 1 1
cf. Orfelia sp. 2 (not captured) 21 –
unidentified species (not captured) 25 –
Sciaridae spp. 16 1
S. streptopoides Mycetophilidae Brevicornu sp. 2 25 0.92
Phronia sp.1 4 0.5
Zygomyia sp. 2 0.5
unidentified species (not captured) 62 –
Sciaridae spp. 4 0.75
sp. 1 (Fig. 3E), Phronia sp. 1 and Zygomyia sp., and by Sciaridae unidentified nematoceran fly (Table S3). Fungus gnats were the
spp. Other visits were made only occasionally, by an empidid only insects found to carry pollen grains, although the body
fly, cerambycid and chrysomelid beetles, a myrid bug and an pollen counts were not high (about ten on average, Table 1).660 Mochizuki and Kawakita – Fungus gnat pollination and shared floral characters
Table 3. Quantity of nectar secreted per flower during 24 h reproductive success between unicolour (yellow) and bicol-
our (yellow and dark red) variants of fungus gnat-pollinated
Plant No. of samples Nectar volume Lepanthes rupestris, suggesting that dark red pigmenta-
(no. of plants) (mean ± sd, μL)
tion does not increase floral attractiveness. A recent study by
A. japonica (female) 35 (7) 1.64 ± 1.00 Katsuhara et al. (2017) found that the dark red, branched petals
A. japonica (male) 43 (6) 1.02 ± 0.74 of fungus gnat-pollinated Mitella pauciflora do not function as
E. lanceolatus 1 (1) 0.80 ± 0.00 attractants, but rather as footholds. Although it is possible that
E. melananthus 8 (4) 2.13 ± 0.90
E. tricarpus 30 (4) 2.11 ± 1.19
dark colour is mimetic of fungus gnat oviposition sites (e.g.
D. cercidifolius 8 (5) 0.49 ± 0.64 fungal fruiting body, rotten wood) in some species, the availa-
M. fusca 20 (5) – ble information indicates that dark red coloration does not nec-
S. streptopoides 6 (5) 0.82 ± 0.75 essarily function to attract fungus gnats. One possibility is that
dark red reduces the frequency of visits by unwanted visitors
Downloaded from https://academic.oup.com/aob/article-abstract/121/4/651/4781678 by guest on 22 May 2020
such as bees (or hoverflies), which are heavy pollen consumers,
Visitors were observed foraging for nectar (Fig. 3E); in doing because bees are less sensitive to red (Troje, 1993; Briscoe and
so, they collected pollen grains on the femurs of the forelegs. Chittka, 2001; Lunau et al., 2011). Besides fungus gnat-polli-
nated plants, dark red flowers are common among plants with
specialized dipteran pollinators, such as those that attract sap-
rophagous, coprophagous or mycophagous flies (van der Niet et
Nectar secretion
al., 2011; Chen et al., 2015; Policha et al., 2016) or gall midges
Nectar secretion was confirmed in all species except M. fusca. (Luo et al., 2017) for pollination. Behavioural experiments and
Although the floral visitors to the flowers of M. fusca showed study of dipteran vision will help us understand the ecological
nectar-feeding behaviour, the nectar could not be sampled by role of dark red pigmentation in pollination. For example, some
the method we used, probably due to the very small amount of saprophytic flies have innate preference for dark colours under
nectar produced by the flowers of this species. The amount of the presence of oligosulphides (Chen et al., 2015).
nectar accumulated in 24 h ranged from 0.49 µL in D. cercidi- Unlike floral colour, floral scent is hypothesized to play a
folius to 2.13 µL in E. melananthus (Table 3). critical role in fungus gnat-pollination systems (Vogel and
Martens, 2000; Okamoto et al., 2015; Katsuhara et al., 2017).
For example, some plants pollinated by fungus gnats (Arisaema,
DISCUSSION Corybas and Heterotropa) are thought to employ brood site
mimicry because they produce a mushroom-like scent and the
Our results indicate that the most important pollinator group eggs and larvae of fungus gnats have been observed inside their
amongst all plant studies was fungus gnats (Table 1, Fig. 4). In flowers [Sugawara, 1988; Vogel and Martens, 2000; Kelly et al.,
some of the plant species studied, fungus gnats did not have the 2013; Woodcock et al., 2014; but Kelly and Gaskett (2014) and
largest pollen load (e.g. A. japonica, E. tricarpus and M. fusca), Kuiter and Findlater-Smith (2017) refuted fungal mimicry in
or the pollen load was small (e.g. S. streptopoides). However, Corybas]. The studied plants also produced floral scents that
because fungus gnats were by far the most frequent visitors, are detectable to the human nose, which resembled those of
their overall contribution to pollination was higher than that fermented dairy products (e.g. yogurt or fresh cheese) or the
of other insects (Table 1; Ackerman and Mesler, 1979; Mesler leaves of fish mint (Houttuynia cordata) (Table 4). However,
et al., 1980). In other species, the numbers of pollen grains car- no fungus gnat visitors exhibited oviposition behaviour, and no
ried by fungus gnats were comparable to or higher than those insect eggs were found on the flowers, suggesting that brood
of other insects (E. melananthus, E. lanceolatus and D. cercid- site mimicry is not involved. Most nematoceran flies, includ-
ifolius; Table 1). In these cases, the PIIs of fungus gnats were ing fungus gnats, feed on nectar as adults (Larson et al., 2001),
greater than 0.85, suggesting their high effectiveness (Table 1, so our observation that the studied plants produce floral nectar
Fig. 4). The PIIs of fungus gnats were low in A. japonica and further indicates that the system is reward-based. Further study
E. tricarpus (0.53 and 0.55, respectively), so these two spe- of floral scent chemistry and the natural history of fungus gnat
cies may have more generalized pollination systems than other pollinators may reveal how the plant species examined in the
studied species. Although other insects may also contribute to current study attract fungus gnats and why the composition of
pollination, especially in A. japonica and E. tricarpus, fungus fungus gnat visitors was female-biased (Table 4).
gnats probably provide the largest contribution to the pollin- In addition to floral colour and odour, overall floral architec-
ation of all the studied plant species. ture may also be important in facilitating pollination by fun-
A notable characteristic shared among the plants studied gus gnats. Fungus gnats generally have very short mouthparts
is their dark red floral pigmentation, but the ecological role that restrict their foraging to open flowers with exposed nec-
of such a floral display is currently unknown. Dark red floral tar (Larson et al., 2001; Fig. 3F–I). The fungus gnats observed
pigmentation is also present in 17 of the 20 previously known in this study consumed nectar by pressing the head and tho-
fungus gnat-pollinated genera (Table S1), although greenish rax onto the flat floral surface, promoting contact between the
flowers, such as those of S. streptopoides, are also common. hairy femur and head and the nearly sessile stamens and pistils
In general, fungus gnats are attracted to bright colours such (Fig. 3C–E). This foraging behaviour results in the attachment
as yellow, which is used to trap sciarid flies in greenhouses of a massive pollen load to the ventral surface of the fungus gnat
(Cloyd and Dickinson, 2005). A study conducted by Tremblay body (Fig. 3F–I; Table 1). In contrast, insects with long mouth-
and Ackerman (2007) showed that there is no difference in parts can consume nectar without coming into contact with theMochizuki and Kawakita – Fungus gnat pollination and shared floral characters 661
Table 4. Summary of fungus gnat pollination in the five studied genera
Genus Aucuba Euonymus Disanthus Micranthes Streptopus
Species A. japonica E. melananthus, D. cercidifolius M. fusca S. streptopoides
E. lanceolatus,
E. tricarpus
Family Garryaceae Celastraceae Hamamelidaceae Saxifragaceae Liliaceae
Order Garryales Celastrales Saxifragales Saxifragales Liliales
Inflorescence type Compound raceme Cyme A pair of adpressed Raceme Uniflowered
flowers
Floral colour Dark-red Dark-red Dark-red Dark-red to pinkish Greenish white with dark-
red markings
Filament662 Mochizuki and Kawakita – Fungus gnat pollination and shared floral characters
Saunders, 2015) and Osyris (Santalaceae; Aronne et al., 1993). Fuller G. 1994. Observations on the pollination of Corybas “A”. New Zealand
Pollination by fungus gnats may have been overlooked because Native Orchid Group Journal 52: 18–22.
Gaskett AC, Endler JA, Phillips RD. 2017. Convergent evolution of sexual
many fungus gnats are nocturnal and both flowers and pollina- deception via chromatic and achromatic contrast rather than colour mim-
tors are inconspicuous. Previously, floral characteristics that are icry. Evolutionary Ecology 31: 205–227.
predictive of fungus gnat pollination were unknown, which also Goldblatt P, Bernhardt P, Vogan P, Manning JC. 2004. Pollination by fun-
limited the discovery of new examples. The set of floral charac- gus gnats (Diptera: Mycetophilidae) and self-recognition sites in Tolmiea
menziesii (Saxifragaceae). Plant Systematics and Evolution 244: 55–67.
teristics we found that is shared among fungus gnat pollinated- Hansen JE, Molau U. 1994. Pollination biology, mating system, and seed set
plants may facilitate the discovery of other examples of fungus in a Danish population of Saxifraga granulata. Nordic Journal of Botany
gnat pollination. 14: 257–268.
Heiduk A, Brake I, von Tschirnhaus M, et al. 2016. Ceropegia sandersonii
mimics attacked honeybees to attract kleptoparasitic flies for pollination.
SUPPLEMENTARY DATA Current Biology 26: 2787–2793.
Downloaded from https://academic.oup.com/aob/article-abstract/121/4/651/4781678 by guest on 22 May 2020
Inouye DW, Larson BMH, Ssymank A, Kevan PG. 2015. Flies and flowers
Supplementary data are available online at www.aob.oxford- III: ecology of foraging and pollination. Journal of Pollination Ecology
journals.org and consist of the following. Table S1: Summary 16: 115–133.
Johnson SD, Schiestl FP. 2016. Floral mimicry. Oxford: Oxford University Press.
of known fungus gnat-pollinated genera. Table S2: Summary of Johnson SD, Wester P. 2017. Stefan Vogel’ s analysis of floral syndromes in
field observations. Table S3: List of floral visitors. the South African flora: An appraisal based on 60 years of pollination stud-
ies. Flora 232: 200–206.
Johnson SD, Harris LF, Procheş Ş. 2009. Pollination and breeding systems
ACKNOWLEDGEMENTS of selected wildflowers in a southern African grassland community. South
African Journal of Botany 75: 630–645.
We are grateful to Dr S. Sakai for helpful suggestions Jakovlev J. 2011. Fungus gnats (Diptera: Sciaroidea) associated with dead
throughout the course of the study and Dr W. Toki, Dr wood and wood growing fungi: new rearing data from Finland and
R. Nakadai, S. Furukawa, T. Hirano, Y. Higuchi and K. Takeda Russian Karelia and general analysis of known larval microhabitats in
Europe. Entomologica Fennica 22: 157–189.
for discussion. We also thank Drs J. Karron, J. Ollerton and Jakovlev J. 2012. Fungal hosts of mycetophilids (Diptera: Sciaroidea exclud-
R. Phillips for providing valuable comments on an earlier draft ing Sciaridae): a review. Mycology 3: 11–23.
of the manuscript. This work was supported by a Grant-in-Aid Jones DL. 1974. The pollination of Acianthus caudatus R. Br. Victorian
for JSPS Fellows to K.M. (grant No. 15J0117) and a Grant-in- Naturalist 91: 272–274.
Aid for Scientific Research (B) to A.K. (grant No. 15H04421). Katsuhara KR, Kitamura S, Ushimaru A. 2017. Functional significance of
petals as landing sites in fungus-gnat pollinated flowers of Mitella pauci-
flora (Saxifragaceae). Functional Ecology 31: 1193–1200.
Kearns CA. 2001. North American dipteran pollinators: assessing their value
LITERATURE CITED and conservation status. Conservation Ecology 5: 5.
Abe T. 2001. Flowering phenology, display size, and fruit set in an understory Kearns CA. Inouye DW. 1993. Techniques for pollination biologists. Boulder,
dioecious shrub, Aucuba japonica (Cornaceae). American Journal of CO: University Press of Colorado.
Botany 88: 455–461. Kelly MM, Gaskett AC. 2014. UV reflectance but no evidence for colour mim-
Ackerman J, Mesler M. 1979. Pollination biology of Listera cordata icry in a putative brood-deceptive orchid Corybas cheesemanii. Current
(Orchidaceae). American Journal of Botany 66: 820–824. http://doi. Zoology 60: 104–113.
org/10.2307/2442469 Kelly MM, Toft RJ, Gaskett AC. 2013. Pollination and insect visitors to the
Aronne G, Wilcock CC, Pizzolongo P. 1993. Pollination biology and sexual putatively brood-site deceptive endemic spurred helmet orchid, Corybas
differentiation of Osyris alba (Santalaceae) in the Mediterranean region. cheesemanii. New Zealand Journal of Botany 51: 155–167.
Plant Systematics and Evolution 188: 1–16. King C, Ballantyne G, Willmer PG. 2013. Why flower visitation is a poor
Averyanov LV, Endress PK, Quang BH, Nguyen KS, Nguyen DV. 2017. proxy for pollination: measuring single-visit pollen deposition, with impli-
Disanthus ovatifolius (Hamamelidaceae), a new species from northwest- cations for pollination networks and conservation. Methods in Ecology
ern Vietnam. Phytotaxa 308: 104–110. and Evolution 4: 811–818.
Barriault I, Barabé D, Cloutier L, Gibernau M. 2010. Pollination ecology Kuiter RH, Findlater-Smith MJ. 2017. Initial observations on the pollination
and reproductive success in Jack-in-the-pulpit (Arisaema triphyllum) in of Corybas (Orchidaceae) by fungus-gnats (Diptera: Sciaroidea). Aquatic
Québec (Canada). Plant Biology 12: 161–171. Photographics 5: 1–19.
Blanco MA, Barboza G. 2005. Pseudocopulatory pollination in Lepanthes Larson B, Kevan P, Inouye D. 2001. Flies and flowers: taxonomic diversity
(Orchidaceae: Pleurothallidinae) by fungus gnats. Annals of Botany 95: of anthophiles and pollinators. The Canadian Entomologist 133: 439–465.
763–772. Lefebvre V, Fontaine C, Villemant C, Daugeron C. 2014. Are empidine
Briscoe AD, Chittka L. 2001. The evolution of color vision in insects. Annual dance flies major flower visitors in alpine environments? A case study in
Review of Enthomology 46: 471–510. the Alps, France. Biology Letters 10: 20140742.
Chen G, Ma XK, Jürgens A, et al. 2015. Mimicking livor mortis: a well- Lindsey AH. 1984. Reproductive biology of Apiaceae. I. Floral visitors to
known but unsubstantiated color profile in sapromyiophily. Journal of Thaspium and Zizia and their importance in pollination. American Journal
Chemical Ecology 41: 808–815. of Botany 71: 375–387.
Cloyd RA, Dickinson A. 2005. Effects of growing media containing diato- Liston A. 2003. A new interpretation of floral morphology in Garrya
maceous earth on the fungus gnat Bradysia sp. nr. coprophila (Litner) (Garryaceae). Taxon 52: 271–276.
(Diptera: Sciaridae). Hortscience 40: 1806–1809. Lunau K, Papiorek S, Eltz T, Sazima M. 2011. Avoidance of achromatic
Dahling GV. 1978. Systematics and evolution of Garrya. Contributions from colours by bees provides a private niche for hummingbirds. Journal of
the Gray Herbarium of Harvard University 209: 1–104. Experimental Biology 214: 1607–1612.
Duque-Buitrago CA, Alzate-Quintero NF, Otero JT. 2014. Nocturnal pol- Luo SX, Liu TT, Cui F, Yang ZY, Hu XY, Renner SS. 2017. Coevolution
lination by fungus gnats of the colombian endemic species, Pleurothallis with pollinating resin midges led to resin-filled nurseries in the androecia,
marthae (orchidaceae: Pleurothallidinae). Lankesteriana 13: 407–417. gynoecia and tepals of Kadsura (Schisandraceae). Annals of Botany 120:
Fenster CB, Armbruster WS, Wilson P, Dudash MR, Thomson JD. 2004. 653–664.
Pollination syndromes and floral specialization. Annual Review of Ecology Ma SJ. 2001. A revision of Euonymus (Celastraceae). Thaiszia 11: 1–264.
and Systematics 35: 375–403. Mar SS, Saunders RMK. 2015. Thismia hongkongensis (Thismiaceae): a new
Fægri K, van der Pijl L. 1979. The principles of pollination ecology, 3rd edn. mycoheterotrophic species from Hong Kong, China, with observations on
Oxford: Pergamon Press. floral visitors and seed dispersal. Phytokeys 33: 21–33.Mochizuki and Kawakita – Fungus gnat pollination and shared floral characters 663
Mayfield MM, Waser NM, Price MV. 2001. Exploring the “most effective Spears EE. 1983. A direct measure of pollination effectiveness. Oecologia 57:
pollinator principle” with complex flowers: bumblebees and Ipomopsis 196–199.
aggregata. Annals of Botany 88: 591–596. Ssymank A, Kearns CA, Pape T. 2008. Pollinating flies (Diptera): a major
Mesler MR, Ackerman JD, Karen LL. 1980. The effective ness of fungus contribution to plant diversity and agricultural production. Biodiversity 9:
gnats as pollinators. American Journal of Botany 67: 564–567. 86–89.
Mesler M, Bencie R, Hayashi B. 2010. A resurrection for Siskiyou bells, Steffan WA. 1981. Sciaridae. In: McAlpine JF, Peterson BV, Shewell GE,
Prosartes parvifolia (Liliaceae), a rare Siskiyou mountains endemic. Teskey HJ, Vockeroth JR, Wood DM, eds. Manual of Nearctic Diptera,
MADROÑO 57: 129–135. vol. 1. Ottawa: Research Branch Agriculture Canada, Monograph 27,
Moldenke AR. 1976. California pollination ecology and vegetation types. 247–256.
Phytologia 34: 305–361. Stökl J, Brodmann J, Dafni A, Ayasse M, Hansson BS. 2011. Smells
Ne’eman G, Jürgens A, Newstrom-Lloyd L, Potts SG, Dafni A. 2010. A like aphids: orchid flowers mimic aphid alarm pheromones to attract
framework for comparing pollinator performance: effectiveness and effi- hoverflies for pollination. Proceedings of the Royal Society B 278:
ciency. Biological Reviews 85: 435–451. 1216–1222.
van der Niet T, Hansen DM, Johnson SD. 2011. Carrion mimicry in a South Stull GW, de Stefano RD, Soltis DE, Soltis PS. 2015. Resolving basal lamiid
Downloaded from https://academic.oup.com/aob/article-abstract/121/4/651/4781678 by guest on 22 May 2020
African orchid: flowers attract a narrow subset of the fly assemblage on phylogeny and the circumscription of Icacinaceae with a plastome-scale
animal carcasses. Annals of Botany 107: 981–992. data set. American Journal of Botany 102: 1794–1813.
Oelschlägel B, Nuss M, von Tschirnhaus M, et al. 2014. The betrayed thief: Suetsugu K. Sueyoshi M, 2017. Subterranean flowers of Aspidistra elatior are
the extraordinary strategy of Aristolochia rotunda to deceive its pollina- mainly pollinated by not terrestrial amphipods but fungus gnats. Ecology,
tors. New Phytologist 206: 342–351. doi:10.1002/ecy.2021.
Okamoto T, Okuyama Y, Goto R, Tokoro M, Kato M. 2015. Parallel chemi- Sugawara T. 1988. Floral biology of Heterotropa tamaensis (Aristolochiaceae)
cal switches underlying pollinator isolation in Asian Mitella. Journal of in Japan. Plant Species Biology 3: 7–12.
Evolutionary Biology 28: 590–600. Teichert H, Dötterl S, Frame D, Kirejtshuk A, Gottsberger G. 2012.
Okuyama Y, Kato M, Murakami N. 2004. Pollination by fungus gnats in A novel pollination mode, saprocantharophily, in Duguetia cadav-
four species of the genus Mitella (Saxifragaceae). Botanical Journal of the erica (Annonaceae): a stinkhorn (Phallales) flower mimic. Flora 207:
Linnean Society 144: 449–460. 522–529.
Okuyama Y, Pellmyr O, Kato M. 2008. Parallel floral adaptations to pollina- The Plant List. 2013. Version 1.1. Published on the Internet; http://www.the-
tion by fungus gnats within the genus Mitella (Saxifragaceae). Molecular plantlist.org/ (accessed 1 September 2017).
Phylogenetics and Evolution 46: 560–575. Tremblay RL, Ackerman JD. 2007. Floral color patterns in a tropical orchid:
Ollerton J. 2017. Pollinator diversity: distribution, ecological function, and conserva- Are they associated with reproductive success? Plant Species Biology 22:
tion. Annual Review of Ecology, Evolution, and Systematics 48: 353–376. 95–105.
Ollerton J, Dötterl S, Ghorpadé K, et al. 2017. Diversity of Diptera families Troje N. 1993. Spectral categories in the learning behaviour of blowflies.
that pollinate Ceropegia (Apocynaceae) trap flowers: an update in light of Zeitschrift für Naturforschung C 48: 96–104.
new data and phylogenetic analyses. Flora 234: 233–244. Vockeroth JR. 1981. Mycetophilidae. In: McAlpine JF, Peterson BV, Shewell
Orford KA, Vaughan IP, Memmott J. 2015. The forgotten flies: the impor- GE, Teskey HJ, Vockeroth JR, Wood DM, eds. Manual of Nearctic
tance of non-syrphid Diptera as pollinators. Proceedings of the Royal Diptera, vol. 1. Ottawa: Research Branch Agriculture Canada, Monograph
Society B 282: 20142934. 27, 223–246.
Policha T, Davis A, Barnadas M, Dentinger BTM, Raguso RA, Roy BA. Vogel S. 1973. Fungus gnat flowers and fungus mimesis. In: Brantjes NBM,
2016. Disentangling visual and olfactory signals in mushroom-mimicking Linskens HF, eds. Pollination and dispersal. Nijmegen: University of
Dracula orchids using realistic three-dimensional printed flowers. New Nijmegen, 13–18.
Phytologist 210: 1058–1071. Vogel S. 1978. Pilzmückenblumen und Pilzmimeten. Flora 167: 329–398.
Proctor M, Yeo P, Lack A. 1996. The natural history of pollination. Portland, Vogel S, Martens J. 2000. A survey of the function of the lethal ket-
OR: Timber Press. tle traps of Arisaema (Araceae), with records of pollinating fungus
Phillips RD, Scaccabarozzi D, Retter BA, et al. 2014. Caught in the gnats from Nepal. Botanical Journal of the Linnean Society 133:
act: Pollination of sexually deceptive trap-flowers by fungus gnats in 61–100.
Pterostylis (Orchidaceae). Annals of Botany 113: 629–641. Willmer P. 2011. Pollination and floral ecology. Princeton, NJ: Princeton
Ren ZX, Li DZ, Bernhardt P, Wang H. 2011. Flowers of Cypripedium University Press.
fargesii (Orchidaceae) fool flat-footed flies (Platypezidae) by faking fun- Whittall JB, Hodges SA. 2007. Pollinator shifts drive increasingly long nectar
gus-infected foliage. Proceedings of the National Academy of Sciences of spurs in columbine flowers. Nature 447: 706–709.
the United States of America 108: 7478–7480. Woodcock TS, Larson BMH, Kevan PG, Inouye DW, Lunau K. 2014.
Sakai S, Inoue T. 1999. A new pollination system: dung-beetle pollination Flies and flowers ii: floral attractants and rewards. Journal of Pollination
discovered in Orchidantha inouei (Lowiaceae, Zingiberales) in Sarawak, Ecology 12: 63–94.
Malaysia. American Journal of Botany 86: 56–61. Xiao Y, Neog B, Xiao Y, Li X, Liu J, He P. 2009. Pollination biology of
Scanlen E. 2006. Flies’ eggs in Nematoceras triloba agg. New Zealand Native Disanthus cercidifolius var. longipes, an endemic and endangered plant in
Orchid Group Journal 98: 34. China. Biologia 64: 731–736.
Shuttleworth A, Johnson SD, Jürgens A. 2017. Entering through the narrow Yamashiro T, Yamashiro A, Yokoyama J, Maki M. 2008. Morphological
gate: a morphological filter explains specialized pollination of a carrion- aspects and phylogenetic analyses of pollination systems in the Tylophora-
scented stapeliad. Flora 232: 92–103. Vincetoxicum complex (Apocynaceae-Asclepiadoideae) in Japan. Biological
Søli GEE, Vockeroth RJ. Matile L. 2000. A.4. Families of Sciaroidea. In: Journal of the Linnean Society 93: 325–341.
Papp L, Darvas B, eds. Contribution to a manual of Palaearctic Diptera. Young A. 1985. Studies of cecidomyiid midges (Diptera: Cecidomyiidae) as
Budapest: Science Herald, Appendix, 49–92. cocoa pollinators (Theobroma cacao L) in Central America. Proceedings
Song B, Chen G, Stöcklin J, et al. 2014. A new pollinating seed-consum- of the Entomological Society of Washington 87: 49–79.
ing mutualism between Rheum nobile and a fly fungus gnat, Bradysia Zych M, Goldstein J, Roguz K. 2013. The most effective pollinator revisited:
sp., involving pollinator attraction by a specific floral compound. New pollen dynamics in a spring-flowering herb. Arthropod-Plant Interactions
Phytologist 203: 1109–1118. 7: 315–322.You can also read
